EP0202536A2

EP0202536A2 - Signal processing circuit for a color video printer

Info

Publication number: EP0202536A2
Application number: EP86106191A
Authority: EP
Inventors: Yasunori Kobori; Toshihiko Gotoh; Kentaro Hanma
Original assignee: Hitachi Ltd
Current assignee: Hitachi Ltd
Priority date: 1985-05-09
Filing date: 1986-05-06
Publication date: 1986-11-26
Also published as: EP0202536A3; JPS61256886A; US4734759A

Abstract

A signal processing circuit for a color video printer according to the present invention comprises first and second memories (302, 202), data interpolation circuit (401) and printer. The first memory (302) stores a color signal which requires the largest amount of information among signals of three primary colors in a video signal. The second memory (303) stores the other two color signals sampled alternately.

The data interpolation circuit (401) reproduces the color signals of lacking parts by the use ofthe color signals stored in the second memory (303).

The color signals delivered from the first memory (302) or the data interpolation circuit (401) are successively sent to the printer so as to be printed.

Description

BACKGROUND OF THE INVENTION:

The present invention relates to a signal processing circuit for a color video printer which depicts a color print close in quality to a photograph on a printing paper on the basis of a color video signal.
Regarding prior-art color video printers, 'IEEE Transactions on Consumer Electronics, Vol. CE-28, No. 3, August 1982' contains a paper entitled "Color Video Picture Printer" by Sohei Masuda, and 'IEEE Transactions on Consumer Electronics, Vol. CE-31, No. 3, August 1985' contains a paper entitled "A COLOR VIDEO PRINTER WITH SUBLIMITATION DYE TRANSFER METHOD". by Yasunori Kobori et al. who are the inventors of the present invention.
As stated in the papers, with the prior-art video printers, a video signal of one frame is disjoined into R (red), G (green) and B (blue) signals which are the components of a color signal or into Cy (cyan), Ye (yellow) and Mg (magenta) signals which correspond to the color components of a coating material. The resulting signals are respectively converted into digital signals, which are once stored in dedicated semiconductor frame memories.
The respective color signals are read out and printed in a predetermined order. In this case, the frame memories and analog-to-digital converters (hereinbelow, termed A/D converters) for converting the color signals are driven at an identical clock frequency. Therefore, all the video signals can be uniformly stored in the frame memories within the frame period thereof (33 msec, in the NTSC format, and 40 msec. in the PAL format).
Since, however, each of the prior-art video printers requires the A/D converters and the frame memories for the respective color signals, the circuit arrangement thereof becomes complicated. It also has a problem in the aspect of cost because a . large number of RAMs (random access memories) are needed as the frame memories, and the A/D converters are needed in the number of three.

SUMMARY OF THE INVENTION:

An object of the present invention is to reduce the memory capacity of memory means for storing video signals.
Another object of the present invention is to provide a signal processing circuit for a video printer which has a simple circuit arrangement.
The color video printer of the present invention comprises a first memory for storing the first color signal which requires the largest amount of information among the three color signals of a color video signal, and a second memory for storing the remaining second and third color signals. The color signal to be stored in the second memory is input at a sampling rate lower than that of the color signal to be stored in the first memory.
The color video printer further comprises an interpolation circuit by which the color signal stored in the second memory is interpolated to reproduce data.

BRIEF DESCRIPTION OF THE DRAWINGS:

Figs. la and lb are block diagrams each showing an embodiment of a signal processing circuit for a video printer according to the present invention;
Fig. 2a is an explanatory diagram of-an output singal from a synthesis switch in Fig. la or lb, while Figs. 2b and 2c are explanatory diagrams of digital color signals respectively separated from the signal shown in Fig. 2a;
Fig. 3 is an explanatory diagram showing a printing operation in fig. la or-lb;
Fig. 4 is a block diagram showing a practicable example of a digital interpolation cirucit in Fig. la or Ib;
Figs. 5a and 5b are schematic diagrams showing digital color signals before and after the data interpolation of the practicable example respectively;
Figs. 6a and 6b and Figs. 7a and 7b are explanatory diagrams showing the data interpolation operations in Fig. 4 respectively;
Fig. 8 and Fig. 9 are block diagrams each showing another practicable example of the digital interpolation circuit in Fig. la or lb;
Fig. 10 is an explanatory diagram showing the operation of the practicable example shown in Fig. 9;
Figs. 11 and 12 are block diagrams each showing still another practicable example of the digital interpolation circuit in Fig. la or lb;
Fig. 13 is an explanatory diagram showing the positional relations between lacking parts and picture element parts in the practicable example of Fig. 12;
Fig. 14 is an explanatory diagram showing a contour correction effect in the practicable example of Fig. 12; and
Fig. 15 is an explanatory diagram showing a method of detecting the change of a digital G-signal in the practicable example of Fig. 12.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS:

Now, embodiments of the present invention will be described with reference to the drawings.
Fig. la is a block diagram showing an embodiment of a signal processing circuit for a video printer according to the present invention.
The signal processing circuit of the video printer in this embodiment is constructed of a signal input portion 1 which is supplied with R, G and B signals, an A/D conversion portion 2 which digitizes the color signals, a memory portion 3 which stores the respective digital color signals, a digital processing portion 4 which processes and selects respective signals read out of the memory portion 3, a signal conversion portion 5 which converts color signals from the digital - processing portion 4 into signals required for printing and supplies the latter to a head assembly 6, and a system controller 7 which controls the above blocks.
The signal input portion 1 consists of an R-signal input terminal 101 to which the R signal is input, a G-signal input terminal 102 to which the G signal is input, and a B-signal input terminal 103 to which the B signal is input.
The A/D conversion portion 2 is constructed of A/ D converters 201, 202 and 203.
The memory portion 3 consists of a frame memory 302 which stores the G signal, a frame memory 303 which stores the R signal and the B signal, a synthesis switch 301 by which the signals to be stored in the frame memory 303 are switched, and a memory controller 304 which controls the synthesis switch 301, the frame memories 302, 303, the A/ D converters 201, 202, 203, and a digital interpolation circuit 401.
The digital processing portion 4 is constructed of the digital data interpolation circuit (hereinbelow, termed "digital interpolation circuit) 401 which synthesizes the R signal and B signal on the basis of the signals from the frame memory 303 of the memory portion 3, and a signal selector 402 by which the R signal or B signal from the digital interpolation circuit 401 and the G signal from the frame memory 302 are output in the order of printing.
The signal conversion portion 5 consists of a one-line memory 501, a pulse generator 502 and a half-tone controller 503. The pulse generator 502 converts data stored in the one-line memory 501, into a signal of pulse width corresponding to a half-tone control signal from the half-tone controller 503 and supplies the latter to the head assembly 6.
Next, the operation of this embodiment will be described.
The G signal, R signal and B signal are respectively input to the input terminals 101, 102 and 103 of the signal input portion 1, and they are supplied to the A/D conversion portion 2. In the A/D conversion portion 2, these color signals are converted into the digital signals of, e.g., 6 bits at the same timing by the A/ D converters 201, 202 and 203 which are controlled by the memory controller 304 of the memory portion 3, whereupon the digital signals are supplied to the memory portion 3. In the memory portion 3, the G signal digitized by the A/D converter 201 (hereinbelow, termed "digital G-signal E_DG") is immediately written into the frame memory 302. On the other hand, the R signal digitized by the A/D converter 202 (hereinbelow, termed "digital R-signal E_DR") and the B signal digitized by the A/D converter 203 (hereinbelow, termed "digital B-signal E_DB") are written into the common frame memory 303 through the synthesis switch 301.
Here, letting E_Y denote the intensity of a luminance signal, and E_R' EG and E_B denote the intensities of the R, G and B color signals respectively, the following holds:
This indicates that the luminance signal E is obtained by mixing 30% of the R color signal E_R, 59% of the G color signal and 11% of the B color signal.
Accordingly, when the amount of information of the G color signal EG necessary for reproducing the luminance signal Ey is expressed by 1 (one), the amount of information of the R color signal E_R may be about 1/2, and that of the B color signal E_B may be as small as about 1/5.
In the present invention, therefore, the G color signal EG is A/D-converted at a sampling frequency f_G so as to store the digital signal in the memory 302, whereas the R or B color signal E_R or E_B is A/D-converted at a sampling frequency ½ of f_G so as to store the digital signal in the frame memory 303.
Concretely, the A/ D converters 201, 202 and 203 A/D-converts the G, R and B color signals E_G, E_R and E_B supplied from the corresponding terminals 101, 102 and 103 at the identical frequency fG, respectively. Since the resulting digital R signal E_DR and digital B signal E_DB are alternately sampled by the synthesis switch 301 and are then written into the frame memory 303, the R and B color signals are, in effect, A/D-converted at the frequency ½^f _G.
Alternatively, the analog R color signal E_R and B signal E_Bmay well be alternately input to a common A/D converter as shown in Fig. lb.
Referring to Fig. lB, the input terminals 102 and 103 are connected to the input terminals of a synthesis switch 305. The output end of the synthesis switch 305 is connected to the input end of the common A/D converter 204, the output end of which is connected to the frame memory 303.
Since the R color signal E_R and B color signal E_B applied to the respective input terminals 102 and 103 are alternately supplied to the A/D converter 204 by the synthesis switch 305, the digital R signal E_DR and digital B signal E_DB are alternately stored in the frame memory 303. Also in the circuit shown in . Fig. lb, the R color signal E_R of B color signal E_B is sampled at the frequency equal to half of that of the G color signal E_G as in the circuit shown in Fig. lA.
Fig. 2a shows data items which are to be stored in the frame memory 303. ® indicates the digital R signal E_DR, and Ⓑ the digital B signal E_DB. The digital R signal E_DR and the digital B signal E_DB are stored dot-sequentially. (The signal in which the digital R signal E_DR and the digital B signal E_DB appear dot-sequentially is called "R/B signal".)
At this time, the amount of information of each of the digital R and B signals E_DR, E_DB which constitute the R/B signal is ½ of that of the digital G signal E_DG. Since, however, the sampling points of the R/B signal and those of the digital G signal are equal in number, the memory capacity of the frame memory 303 for storing the R/B signal is equal to that of the frame memory 302 for storing the digital _G signal EDGe
On the basis of this fact, the embodiment reduces the memory capacity of the memory portion to 2/3 times as compared with the prior art in which the frame memories are provided for the respective digital color signals.
When the digital G, R and B signals E_DG, ^E _DR' E_DB for the video signal of one frame have been stored in the frame memories 302 and 303, they are subsequently read out from these frame memories. That is, the digital G signal E_DG is read out from the frame memory 302, and the R/B signal is read out from the frame memory 303, to be supplied to the digital processing portion 4.
Either of the digital R and B signals is extracted from the R/B signal supplied to the digital interpolation circuit 401.
Since the extracted digital R signal E_DR or digital B signal E_DB has been sampled at the sampling frequency equal to1 2 of that of the digital G signal E_DG, the amount of information thereof is insufficient.
The digital R signal E_DR extracted from the R/B by the digital interpolation circuit 401 is shown in Fig. 2b, and the digital B signal E_DB is shown in _Fig. 2c. In these digital color signals ^E _DR and ^E _DB, the signals (hereinbelow, termed "picture elements") of sampling points (hereinbelow, termed "dots") are denoted by ® and Ⓑ, and the dots (hereinbelow, termed "lacking parts") where the picture elements are missed by the synthesis switch 301 are denoted . by marks x. Regarding one frame, as shown in Figs. 2b and 2c, the digital R and G signals E_DR and E_DB extracted from the R/B signal are respectively such that the picture elements ® and the lacking parts x are alternately arrayed and that the picture elements Ⓑ and the lacking parts x are alternately arrayed.
The digital interpolation circuit 401 interpolates the signals of the lacking parts x by the use of the signals of the dots of the actually existing picture elements ® and Ⓑ, thereby to form the digital R and B signals E_DR and E_DB whose information quantities are increased.
The signal selector 402 selects the digital G signal _EDG from the frame memory 302 and the digital R signal E_DR or B signal E_DB from the digital interpolation circuit 401 in succession in the order of printing and supplies them to the signal conversion portion 5.
First, in order to print a green image, the digital G signal E_DG is selected, and the signals thereof corresponding to one frame are supplied to the signal conversion portion 5. Subsequently, in order to print a red or blue image, the output signal of the digital interpolation circuit 401 is selected.. In the digital processing portion 4, in the period during which the digital interpolation circuit 401 is selected, this circuit 401 extracts the digital R signal E_DR from the R/B signal in order to first print red. The digital R signal E_DR extracted and subjected to the interpolation processing is supplied to the signal conversion portion 5 by the signal selector 402. In order to subsequently print blue, the digital interpolation circuit 401 extracts the digital B signal E_DB from the R/B signal. The extracted and interpolated digital R signal E_DR is supplied to the signal conversion portion 5. Such selecting operations of the signal selector 402 are performed under the control of the system controller 7.
In the signal conversion portion 5, the signals of one line from the digital color signal E_DG, E_DG or E_DB input from the signal selector 402 are held in the line memory 501. Under the control of the half-tone controller 503, all the picture elements held in the line memory 501 are simultaneously converted into gradation pulses conforming to the respective information contents by means of the pulse generator 502. At the same time, the gradation pulses are supplied to the head assembly 6 to effect the printing.
Here, the signals corresponding to one line consist of signals which correspond to a column of picture elements arrayed in a direction perpendicular to the horizontal scanning direction of a television screen. As illustrated in Fig. 3, the gradation pulses obtained by converting the signals of the picture elements of a line signal are simultaneously applied to the respective heads 6a, 6b, ..... of the head assembly 6 which are arranged in the vertical direction (as viewed on the drawing) of printing paper P, whereby the information is printed on the printing paper P.
In Fig. 3, the moving direction of the head assembly 6 relative to the printing paper P is as indicated by an arrow. In the signal conversion portion 5, the digital color signals of one line are held in the line memory 501 every frame of the applied digital color signal E_DG' ^E _DR or ^E _DB. Accordingly, the head assembly 6 prints line by line rightwards from the left end of the printing paper P as it moves relative to the printing paper P. In addition, the head assembly 6 first prints the green image over the whole area of the printing paper P on the basis of the G signal E_DG, subsequently prints the red image on the basis of the R signal E_DR and further prints . the blue image on the basis of the B signal E_DB, thereby to print one video image on the printing paper P. The above printing order is appointed by the half-tone controller 503 under the control of the system controller 7 shown in Fig. la and Fig. lb.
Although color signals which are actually supplied for the printing are cyan (Cy), yellow (Ye) and magenta (Mg) signals, the R, G and B signals have been used for the description. The Cy, Ye and Mg signals are obtained by converting the R, G and B signals, and the conversion may be executed in the digital processing portion 4 by way of example. Alternatively, color signals to be applied to the input portion 1 may be the Cy, Ye and Mg signals.
As described above, since the digital R signal E_DR and the digital B signal E_DB are stored in the identical memory.303, the memory capacity of the memory portion 3 is reduced. Although the amounts of information of the digital R signal E_DR and B signal E_DB decrease on account of the synthesis switch 301, the lacking information contents are supplemented by the digital interpolation circuit 401. Therefore, the degration of the picture quality of a printed picture can be prevented.
Next, practicable examples of the digital inter-' polation circuit 401 in Fig. la or lb will be described.
Fig. 4 is a block diagram showing one practicable example of the digital interpolation circuit 401. This digital interpolation circuit 401 consists of a signal discriminator 8, a line memory 9, a shift register group 10 which is constructed of shift registers 110 and 111, a mean value circuit 11, a latch circuit 12, switches 13, 14 and 15, an address counter 16, and a pulse generator 17.
Referring to the figure, the R/B signal from the frame memory 303 (Fig. la or 1b) is supplied to the signal discriminator 8, and either digital color signal E_DR or E_DB is extracted. For the sake of explanation, it is assumed here that the digital R signal E_DR be extracted. Then, signals corresponding to one line of the digital R signal E_DR are stored in the line memory 9. This digital R signal for one line (namely, the line signal) E_DR is stored in the line memory 9 in such a manner that, as shown in Fig. 5a, data items corresponding to picture elements ® and data items corresponding to lacking parts
are alternately arrayed in the order of
,
,
,
'
, ......
Here, it is assumed that one line be composed of 512 picture elements and that
be the picture element at the uppermost part of a frame, while
be the part where the picture element
at the lowermost part is missing.
This digital interpolation circuit 401 carries out the data interpolation on the basis of an interpolating method to be stated below.
As illustrated in Fig. 5a, odd-numbered dots have the data of the picture elements, whereas even-numbered dots are the lacking parts and have no data. Now, letting R_2n-1 and R_2n+1 denote the data items (hereinbelow, termed "picture element data") of the respective picture elements
and
, and C_2n denote a data item (hereinbelow, termed "interpolation data") to be interpolated in the lacking part
between them, the following operation is executed:
That is, the mean value of the picture element data of the upper and lower picture elements is used as the interpolation data of the lacking part x_2n. The interpolation based on this operation is executed for all the lacking parts x: This interpolation is not applicable to the last lacking part
. As the interpolation data C₅₁₂ therefor, the picture element- data R₅₁₁ of the picture element
located above is used. That is, the following operation is executed:
Owing to the execution of such interpolation, the digital R signal E_DR for one line becomes as shown in Fig. 5b.
Hereunder, the operation of the practicable example of the digital interpolation circuit 401 (as shown in Fig. 1) in Fig. 4 as based on the above-stated interpolation method will be explained.
The picture element data R_2n+1 and the data C_2n of the lacking part x_2n are read out from the line memory 9 successively in a downward sequence shown in _Fig. 5a, in accordance with address signals from the address counter 16 which is operated by the clock of the memory controller 304 (Fig. la or lb). The address signals produced from the address counter 16 are also supplied to the pulse generator 17. The pulse generator 17 generates switching pulses for switching the switches 13, 14 and 15 in accordance with the address values of the address signals from the address counter 16. When transferring data to the shift register 110, the switch 13 is thrown to the contact c side thereof, so that the picture element data R_2n+1 and the lacking part data x_2n from the line memory 9 are transferred toward the shift registers 110 and 111 through the switch 13 in synchronism with the clock from the memory controller 304. When the output M₁ of the shift register 111 is the picture element data owing to the transfer, the switch 14 is thrown to the contact a side thereof by the switching pulse from the pulse generator 17, and this picture element data item R_2n+1 is supplied to the signal selector 402 (Fig. la or lb) through the switch 14. In contrast, when the output M₁ of the shift register 111 is the lacking part data x_2n, the switch 14 is thrown to the contact b side, and the output M_A of the latch circuit 12 is supplied as the interpolation data C_2n to the signal selector 402 through the switch 14.
This interpolation data item C_2n is formed as follows:

Assuming now that the data of the output M₃ of the line memory 9 be D_2n, the data items of the outputs M₂ and M₁ of the shift registers 110 and 111 are ^D _2n-1 and D_2n-2 respectively. The output M₃ of the line memory 9 and that M₁of the shift register 111 are supplied to the mean value circuit 11, to calculate the mean value of them. Accordingly, the data D_2n-1 of the output M_B of the mean value circuit 11 becomes:

At this time, the latch circuit 12 latches data D_2n-2 one clock time before the mean value circuit 11 delivers the data D_2n-1. Accordingly, the data of the output M_A of the latch circuit 12 is:
In general, when the data of the output M₁ of the shift register 111 is D_2i, the data of the output M_B of the mean value circuit 11 is:
and the data of the output M_A of the latch circuit 12 is:
It is now assumed that, as shown in Fig. 5a, the odd-numbered dots in a certain line be the picture elements
,
,
, ....., while the even-numbered dots be the lacking parts
,
, ...... In addition, picture element data for such picture elements is assumed data D_2i-1 with suffixes of odd numbers, and picture element data for such lacking parts is assumed data D_2i with suffixes of even numbers.
It is assumed that, at a certain point of time, the data of the output M₁ of the shift register 111 be D_2n-1' while the output M₂ and input M₃ of the shift register 110 be D_2n and D_2n+l respectively. Then, the odd-numbered data D_2n-1 of the output M₁ of the shift register 111 is the picture element data and therefore requires no interpolation. The swtich 14 is thrown to the contact a side, and this data item D_2n-1 passes through the switch 14 to be fed into the signal selector 402 (Fig. la or lb). On this occasion, the latch circuit 12 is continually supplied with the output M_B of the mean value circuit 11 for the picture element data items:
through the switch 15 which is thrown on the contact e side thereof. The latch circuit 12, however, still holds the data D2_n-2 [= (D_2n-2 + D_2n)/2] of the preceding output M_Bof the mean value circuit 11.
When the signal selector 402 has received the data D_2n-1 of the output M₁ of the shift register 111, the latch circuit 12 latches the aforementioned data D_2n of the output M_B of the mean value circuit 11. Subsequently, when the shift register 111 is fed with the data D_2n of the output M₂ of the shift register 110 and this shift register 110 is fed with the data D_2n-1 at the input M₃ thereof, the output M₁ of the shift register 111 is the lacking part data x_2n, and hence, the switch 14 is changed-over to the contact b side. Therefore, the data D_2n of the output M_A of the latch circuit 12 is fed into the signal selector 402 through the switch 14.
Since this data item D_2n is the mean value of the picture element data as indicated by the last equation, it becomes the interpolation data C_2n for the 2n-th lacking part
in the line. At this time, the data D_2n+1 of the output M_B of the mean value circuit 11 is:
This is the mean of the two lacking part data items D_2n and D_2n+1.
When the signal selector 402 has received the data D_2n from the latch circuit 12, this latch cir- _cuit 12 latches the aforementioned data D_2n+1 of the output M_B of the mean value circuit 11. Subsequently, the shift register 111 is fed with the data D_2n+1 of the output M₂ of the shift register 110, and this shift register l10 is fed with the data D_2n+2 of its input M₃. Simultaneously therewith, the input M₃ of the shift register l10 becomes data D2n+3, and the switch 14 is thrown to the contact a side.
In this way, when the output M₁ of the shift register 111 is the odd-numbered data and the picture element data R_2n+1, this data item is fed into the signal selector 402, and when it is the even-numbered data and the lacking part data x_2n, the mean value of the picture element data items R_2n and R_2n+1 before and after the lacking part data is fed into the signal selector 402 as the interpolation data C_2n.
In the above, the odd-numbered dots in one line have been the picture elements, and the even-numbered dots have been the lacking parts as shown in Fig. 5a. In the reverse case, as illustrated in Fig. 6b, the switch 14 is thrown to the a side when the output M₁ of the shift register 111 is any of even-numbered data items D_2m-2' D_2m' ....., and it is thrown to the b side when the output M₁ is any of odd-numbered data items D_2m-1, D_2m+1, ...... The other operations are similar to those in the foregoing.
Although the above interpolation is valid for the lacking parts
~
in Fig. 5a, such interpolation data C₅₁₂ cannot be created for the lacking part
. As explained before, the interpolation of this lacking part
is to put the picture element data R₅₁₁ of the picture element
as the interpolation data C512. Next, this interpolation operation will be described with reference to Fig. 7a.
Assuming in Fig. 4 that the input M₃ of the shift register 110 be the picture element data D₅₁₁ of the final picture element
, the data of the output M₁ of the shift register 111 is D_509' and this is supplied to the signal selector 402 (Fig. la or 1b) through the switch 14. Simultaneously, the data D₅₁₀ of the output M_B of the mean value circuit 11 is latched in the latch circuit 12. Here, the data D₅₁₀ is:
Subsequently, when the input M₃ is stored in the shift register 110 and the output M₂ (data D₅₁₀) of this shift register 110 is.stored in the shift register 111, the switch 13 is thrown to the d side thereof. Thus, the data items of the input M₃and output M₂ of the shift register l10 become equal and D_511' On this occasion, the switch 14 is thrown to the b side, and the output M_A of the latch circuit 12 is supplied to the signal selector 402 as the interpolation data of the lacking part
.
Further, when the output M₂ of the data D511 from the shift register 110 is stored in the shift register 111 and the input M₃ of the data D₅₁₁ is stored in the shift register 110, all the data items of the input M₃ and output M₂ of the shift register 110 and the output M₁ of the shift register 111 become D₅₁₁. The switch 14 is thrown to the a side thereof, and the data D₅₁₁ of the output M₁of the shift register 111 is supplied to the signal selector 402. Simultaneously therewith, the data of the output M_B of the mean value circuit 11 becomes the mean value between the data D₅₁₁ of the input M₃of the shift register 110 and the data D₅₁₁ of the output M₁of the shift register 111, that is, the data D₅₁₁ itself, and this is latched in the latch circuit 12.
Further, when the output M₂ of the shift register 110 is stored in the shift register 111 and the input M₃of this shift register 110 is stored therein, the switch 14 is changed-over to the b side thereof, and the output M_A of the data D₅₁₁ from the latch circuit 12 is supplied to the signal selector 402 through the switch 14. This is the interpolation data of the lacking part
in Fi^g. 5a.
Next, a case where the first dot of a line is a lacking part unlike the arrayal of Fig. 5a will be described with reference to Fig. 7b.
When the readout of the line memory 9 is started with the switch 13 thrown to the c side thereof, the data D₁ of the first input M₃of the shift register 110 is lacking part data. When this data item D₁ is stored in the shift register 110, the output M₂ thereof becomes data D₁ and the input M₃ thereof becomes data D₂. At this time, the switch 15 is thrown to the f side thereof, the latch circuit 12 lacthes the data D₂ of the input M₃of the shift register 110. This data item D₂is the first picture element data R₁ in the line. Thus far, the shift register 111 holds indefinite data, and the signal selector 402 is not fed with the output of the digital interpolation circuit 401.
Subsequently, the shift register 111 is fed with the data D₁ of the output M₂of the shift register 110, and the shift register 110 is fed with the data D₂ of the input M₃ thereof. Accordingly, the output M₁ of the shift register 111 becomes the data D₁, and that M₂ of the shift register 110 becomes the data D₂. Besides, the input M₃ of the shift register 110 becomes data D₃. At this time, the data D₂ of the output M_B of the mean value circuit 11 becomes:
and the output M_A of the latch circuit 12 is the data D₂. Further, the switch 15 is changed-over to the e side thereof, but the latch circuit 12 continues to hold the data D₂.
Therefore, the output M₁ of the shift register 111 is the lacking part data D₁, the switch 14 is thrown to the b side, and the data D₂ of the output M_A of the latch circuit 12 is fed into the signal selector 402 through the switch 14 as the first output data of the digital interpolation circuit 401.
Subsequently, when the shift register 111 is fed with the data D₂ of the output M₂ of the shift register 110 and the shift register 110 is fed with the data D₃ of the input M₃ thereof, the switch 14 is changed-over to the a side, and the data D₂ of the output M₁ of the shift register 111 is fed into the signal selector 402 through the switch 14. Simultaneously therewith, the data D₃' of the output M_B of the mean value circuit 11 becomes:
This is latched in the latch circuit 12, but is interrupted by the switch 14:

Thenceforth, operations are similar to those in the foregoing. In this way, the first lacking part - of the line is supplemented by the succeeding picture element.

To sum up the above operations, in the line which consists of, for example, 512 data items held in the line memory 9, intermediate lacking parts (second to 511th data items) are interpolated using the mean values (mean data) of the picture element data items before and after them, and when lacking parts at both the ends of the line (first and 512th data items) need to be interpolated, the second picture element data R₂ is substituted for the first lacking part x₁ and the 511th picture element data R₅₁₁ is substituted for the 512th lacking part x₅₁₂. This method of data interpolation is performed in the direction of the line to be printed, namely, in the vertical direction of the surface of the monitor in Fig. 3.
Fig. 8 shows another practicable example of the digital interpolation circuit 401. This digital interpolation circuit 401 is provided anew with a switch 18 and a data generator 19. The same portions as in the circuit shown in Fig. 4 are assigned identical symbols, and shall not be repeatedly explained.
In the foregoing practicable example shown in Fig. 4, the first or last lacking part of each line has been interpolated with the directly succeeding or directly preceding picture element data respectively. In this practicable example shown in Fig.
8, the first or last dot of each line (in Fig. 5a, the parts of the picture element
or the lacking part
) is interpolated with a predetermined data item, for example, white level data whose bits are all "1". As a result, in the case of the interpolation with the white level data, the number of picture elements per line of a printed image becomes 510 which is about 0.4% smaller than the number of 512, but quite no problem is posed in vision.
Next, the operation of this practicable example will be explained.
Referring to Fig. 8, the switch 18 is thrown to the contact g side thereof as to the second to 511th parts of the line. As in the practicable example shown in Fig. 4, when the output M₁ of the shift register 111 is picture element data R_2n, the switch 14 is thrown to the contact a side, and this data item is fed into the signal selector 402 (Fig. la or lb( through the switches 14 and 18. When the output M₁of the shift register 111 is lacking part data x_2n+1, the switch 14 is thrown to the contact b side, and the output M_A of the latch circuit 12 is fed into the signal selector 402 through the switches 14 and - 18 as interpolation data C_2n+1.
When the output M₁ of the shift register 111 is data R₁or R₅₁₂ expressive of the first of last part of the line, the switch 18 is changed-over to the h side thereof by a switching pulse from the pulse generator 17, and the data of a predetermined value provided by the data generator 19 is fed into the signal selector 402 through the switch 18 as interpolation data C₁ or C₅₁₂. When the data of the predetermined value is assumed the data in which all bits are "1", it is not printed on a printing paper. Accordingly, the uppermost part and lowermost part of each line on a printed frame become a white state when this processing is performed for the digital G, R and B signals.
Thus, with this practicable example, the switches 13 and 15 in Fig. 4 are dispensed with to simplify the circuit arrangement, and to simplify the interpolation processing. Moreover, with the practicable example of Fig. 4, the edges of the printed frame have undesired colors and are conspicuous, whereas with the practicable example of Fig. 8, white edges are formed and are not problematic.
In the above practicable examples, the number of the heads in the head assembly 6 (Fig. la or lb) has- been equal to the number of picture elements per line (the number of 512). However, the number of heads may well be reduced by two into 510 so as to print the second to 511th picture element data in the line by means of the 510 heads. In this case, the first and last dots of the line need not be interpolated, and effects similar to those of the practicable example shown in Fig. 8 are attained. In addition, similar effects are attained when the number of picture elements of one line has two increased into 514 in which the sixth to 513th picture elements are used as data R₆ - R₅₁₃, and the first to 512th parts are subjected to the data interpolation based on the mean values, only these parts being printed.
Fig. 9 shows still another practicable example of the digital interpolation circuit 401 in Fig. 1.
In this practicable example, a line memory portion 9' including line memories 901, 902 and 903, and a latch portion 20 including latch circuits 201, 202 and 203 are added anew to the practicable example of Fig. 4. The same parts as in Fig. 4 are assigned identical symbols, and shall not be repeatedly explained.
Each of the foregoing practicable examples has interpolated the lacking part in the line with the mean value of the picture element data items R_2n-1 and R_2n+1 of the same line, whereas this practicable example interpolates it with the mean value of the picture element data R_2ℓ-1, _2n and R_{2ℓ+1, 2n} of both the adjacent lines.
Next, the operation of this practicable example will be described with reference to Fig. 10.
Now, a line where a lacking part is to be interpolated is assumed the 2k-th line, which is expressed by L_2ℓ. Then, the (2ℓ - l)-th line directly preceding the line L_2ℓ is denoted as L_2ℓ-1, and the (2ℓ + l)-th line directly succeeding the same is denoted as L_2ℓ+1.
In Fig. 9, digital color signals from the signal discriminator 8 are supplied to the line memory portion 9'. In this regard, the line memory portion 9' consists of the three line memories 901, 902 and 903, in which the digital color signals of corresponding lines in a time-sequential order are respectively stored. Here, a line which is stored in the line memory 902 is the line to be fed into the signal selector 402 after the data interpolation. The immediately preceding line is stored in the line memory 903, and the immediately succeeding line in the line memory 901.
Assuming now that the digital color signals of the line L₂₁ be stored in the line memory 902, those of the line L_2ℓ-1 are stored in the line memory 903, and those of the line L_2ℓ+1 are stored in the line memory 901. In addition, data items stored at the same addresses of the line memories 901, 902 and 903 are the data items of those sampling points of the lines stored in the respective memories which are arrayed in the horizontal direction of a frame. Accordingly, when data stored at a certain address of the line memory 902 is the picture element data of the corresponding line, data items stored at the same addresses of the line memories 901 and 903 are the lacking part data items of the corresponding lines.
Now, let's consider a case where the line L_2ℓ is stored in the line memory 902, the 2n-th address of which stores data D_{2ℓ, 2n}. Then, the same addresses of the line memories 903 and 901 store data items D_{2ℓ-1, 2n} and D_{2ℓ+1, 2n}, respectively. In Fig. 10, the lacking part data is indicated by Ⓧ.
When the (2n - 2)-th addresses of the line memories 901, 902 and 903 are appointed by the address counter 16 and data D_2ℓ, _2n-2 being picture element data is read out from the line memory 902 and latched in the latch circuit 202, lacking part data items Ⓧ are read out from the line memories 901 and 903 and are respectively latched in the latch circuits 201 and 203. At this time, the switch 14 is thrown to the a side thereof, and the data D_{2ℓ, 2n-2} of the output M of the latch circuit 202 is fed into the signal selector 402 (Fig. la or lb) through the switch 14. Besides, the output M_B of the mean value circuit 11 is the mean value between the lacking part data items of the outputs of the latch circuits 201 and 203.
Subsequently, when the address counter 16 appoints the next (2n - l)-th address, lacking part data Ⓧ is read out from the line memory 902 and is latched in the latch circuit 202, and data items D_{2ℓ+1, 2n-1} and D_{2ℓ-1, 2n-l} being picture element data items are respectively read out from the line memories 901 and 903 and latched in the latch circuits 201 and 203. Then, the switch 14 is changed-over to the b side thereof, and the mean data of the data items D_{2ℓ+1, 2n-1} and D_{2ℓ-1, 2n-1} delivered as the output M_B of the mean value circuit 11 is fed into the signal selector 402 through the switch 14 as interpolation data D_{2ℓ, 2n-1}.
Subsequently, when the address counter 16 appoints the 2n-th address, data D_{2ℓ, 2n} being picture element data is read out from the line memory 902 and latched in the latch circuit 202, and lacking data items Ⓧ are read out from the line memories 901 and 903. Accordingly, the switch 14 is thrown to the a side, and the data D_{2ℓ, 2n} of the output M_A of the latch circuit 202 is fed into the signal selector 402 through the switch 14.
In this way, when the picture element data D_{2ℓ, 2n} is read out from the line memory 902, it is supplied to the signal selector 402, and when the lacking part data D_{2ℓ, 2n+1} is read out, the mean data of the picture element data items D_{2ℓ+1, 2n+l} and D_{2ℓ-1, 2n-1} read out from the line memories 901 and 903 is supplied to the signal selector 402 as the interpolation data D_{2ℓ, 2n+1}.
When such processing operations for the line L₂₁ have been completed, the line L_2ℓ is transferred from the line memory 902 to the line memory 903, and the line L_2ℓ+1 is transferred from the line memory 901 to the line memory 902. Simultaneously therewith, the line L_2ℓ+2 is supplied from the signal discriminator 8 to the line memory 901. Thus, the line L_2ℓ+1 is similarly processed.
In this manner, the respective lines are successively subjected to the interpolation of the lacking parts and supplied to the signal selector. It is impossible, however, to subject the first line and the last line to such data interpolation of lacking parts. Regarding these lines, therefore, such processing may be performed as the data interpolation along the line as in the embodiment shown in Fig. 4, the substitution of the data expressive of white as in the embodiment shown in Fig. 8, or the check of the printing of these lines as explained before.
Fig. 11 shows still another practicable example of the digital interpolation circuit 401 in Fig. la or lb.
As compared with the practicable example of Fig. 9, this example differs in a latch circuit 10' and a mean value circuit 11'. Portions corresponding to those in Fig. 4 and Fig. 9 are assigned the same symbols.
In this practicable example, interpolation data D_{2ℓ, 2n+1} is formed using the picture element data items R_n-1 and R_n+1 along the line in the practicable example shown in Fig. 4 and the picture element data items D_{2ℓ-1, 2n+1}, D_{2ℓ+1, 2n+1} and D_{2ℓ, 2n+l} of the adjoining lines in the practicable example shown in Fig. 9.
The operation of this practicable example will be described with reference to Fig. 10.
It is now assumed that the lines L_2ℓ+1, L_2ℓ and L_2ℓ-1 be respectively stored in the line memories 901, 902 and 903. Assuming that the data items of the (2n - l)-th addresses of the line memories 901, 902 and 903 appointed by the address counter 16 be respectively latched in the latch circuits 201, 202 and 203, then the data latched in the latch circuit ²⁰¹ is D_{2ℓ+1, 2n-1} and the data latched in the latch circuit 203 is D_{2ℓ-1, 2n-1}, but the data latched in the latch circuit 202 is the lacking part data Ⓧ. Besides, data latched in the latch circuit 10' is data D_{2ℓ, 2n-2} stored at an address [the (2n - 2)-th address] immediately preceding the address in the line memory 902, of the data latched in the latch circuit 202.
When, under such a state, the address counter 16 appoints the next 2n-th address, the data D_{2ℓ, 2n} is read out from the line memory 902. In addition, the lacking part data items Ⓧ are read out from the line memories 901 and 903. Therefore, the mean value circuit 11' creates the mean data D_{2ℓ, 2n-1} among the data D_{2ℓ, 2n} from the line memory 902 and the data items D_{2ℓ+1, 2n-1}, D_{2ℓ-1, 2n-1} and D_{2ℓ, 2n-2} respectively produced by the latch circuits 201, 203 and 10' and provides it as its output M_B. This mean data item D_{2ℓ, 2n-1} is expressed as follows:
At this time, the switch 14 is thrown to the b side, and the data D_{2ℓ, 2n-1} of the output M_B of the mean value circuit 11' is fed into the signal selector 402 (Fig. la or 1b) through the switch 14 as the interpolation data for the lacking part data Ⓧ of the output M_A of the latch circuit 202.
When the interpolation data has been fed, the latch circuits 201, 202, 203 and 10' latch their inputs respectively, and simultaneously, the switch 14 is changed-over to the a side. Thus, the lacking part data items Ⓧ are respectively latched in the latch circuits 201, 203 and 10', but picture element data D_{2ℓ, 2n} is latched in the latch circuit 202.
The data D_{2ℓ, 2n} of the output M_A of the latch circuit 202 is fed into the signal selector 402 through the switch 14.
Thenceforth, in a similar manner, when the output M_A of the latch circuit 202 is the picture element data, it is fed into the signal selector 402 through the switch 14, and when the output M_A is the lacking part data, the mean data among the picture element data from the line memory 902 and the picture element data items from the latch circuits 201, 203 and 10' as created by the mean value circuit 11' is fed into the signal selector 402 as the interpolation data.
In this practicable example, the first and last lacking parts of each line and the first and last line lacking parts cannot be subjected to the data interpolation described above. However, they may be subjected to the data interpolation by the method explained in the embodiment of Fig. 4 or Fig. 9. Alternatively, the first and last lines and the first and last parts of each line may be rendered white on printing paper as explained before.
Fig. 12 shows still another practicable example of the digital interpolation circuit 401 in Fig. la or lb. Added anew are a mean value circuit portion 11" including mean value circuits 115, 116 and 117; a level comparison portion 21 including comparator circuits 210 and 211; data selectors 22 and 23; a shift register portion 24 including shift registers 241, 242 and 243; a latch portion 25 including latch circuits 252 and 253; a latch circuit 26; and a level comparison portion 27 including comparator circuits 270 and 271. The other portions corresponding to those in Fig. la or lb and Fig. 11 are assigned the same symbols.
This practicable example usually performs the data interpolation of lacking parts likewise to the practicable example shown in Fig. 11 and further performs the data interpolation of lacking parts as regards the contour parts of picture elements and so as not to deteriorate them. The contour parts of picture elements and detected with the digital G signal the information content of which is not decreased.
In the example of Fig. 12, the line memory portion 9', the latch portion 20 and the latch circuit 10' operate similarly to those in Fig. ll.
The level comparison portion 11" consists of the mean value circuit 115 which creates the mean data between data read out from the line memory 902 and the output data of the latch circuit 10', the mean value circuit 116 which creates the mean value data between the output data items of the latch circuits 201 and 203, and the mean value circuit 117 which creates the mean value data between the outputs of the mean value circuits 115 and 116. The outputs A_V, A_H and AS of these mean value circuits 115, 116 and 117 are supplied to the data selector 22 as inputs D₁, D₂ and D₀ respectively. Here, the output A_S of the mean value circuit 117 is the same as the output M_B of the mean value circuit 11' in Fig. ll.
In addition, the output data of the latch circuit 10' is supplied to the data selector 23 as an input D₀', and the read-out data of the line memory 902 as an input D₁', and the output data of the latch circuit 201 is supplied thereto as inputs D₂' and D₃'. The inputs are selected according to the outputs of the level comparison portion 27, and the selected data is supplied to the data selector 22 as an input ^D ₃.
Further, in the level comparison portion 21, the output data of the latch circuit 10' and the read-out data of the line memory 902 are compared by the comparator circuit 210, and the output data items of the latch Circuits 201 and 203 are compared by the comparator circuit 211. On the basis of the results of the comparisons, the data selector 22 selects any of the inputs DO - ^D ₃.
Next, the operation of this practicable example will be described with reference to Fig. 13.
It is now assumed that the signal discriminator 8 extract the digital R signal from the R/B signal delivered from the frame memory 303, and that the lines L_2ℓ+1, L_2ℓ and L_2ℓ-1 thereof be respectively stored in the line memories 901, 902 and 903. Letting the picture element data of these lines be R_{i, j} (where i denotes line No., and j denotes dot No.), as shown in Fig. 13, at the (n - l)-th addresses, the line memories 901 and 903 store the lacking part data Ⓧ and the line memory 902 stores picture element data R_{ℓ, n-1}. When note is taken of the n-th addresses, the line memories 901 and 903 store picture element data R_{ℓ+1, n} and R_{ℓ-1, n}, and the line memory 902 stores lacking part data X (which is especially expressed as
. When the (n +1)-th addresses are considered, the line memories 901 and 903 store the lacking part data Ⓧ, and the line memory 902 stores picture element data R_{ℓ, n-1}.
First, when the output M_A of the latch circuit 202 is the picture element data (assumed to be R_{ℓ, n-1}), the switch 14 is thrown to the a side, and this picture element data item R_{ℓ, n-1} is fed into the signal selector 402 through the switch 14. This is the same as in the embodiment shown in Fig. ll. At this time, the line memories 901, 902 and 903 have the next n-th addresses appointed, so that the picture element data items R_{ℓ+1, n} and R_{ℓ-1, n} are respectively read out from the line memories 901 and 903 and that the lacking part data
is read out from the line memory 902.
When the signal selector 402 has been fed with -the picture data R _n-1, the latch circuits 201 - 203 and 10' latch their input data respectively, and the switch 14 is changed-over to the b side. Accordingly, the output of the line memory 902 becomes R_{ℓ, n+1}, and the outputs of the latch circuits 201, 203 and 10' become R_{ℓ+1, n}, R_{k-1, n} and R_{ℓ, n-1} respectively. Simultaneously therewith, the line memories 901, 902 and 903 have the next (n + l)-th addresses appointed, so that the lacking part data items Ⓧ are read out from the line memories 901 and 903 and that the picture element data R_{ℓ, n+1} is read out from the line memory 902. Thus, the output data items A_V, A_H and A_S of the mean value circuits 115, 116 and 117 become as follows respectively:
These are respectively supplied to the data selector 22 as the inputs D₁, D₂ and D₀. Besides, the output Z₂₃ of the data selector 23 is supplied to the data selector 22 as the input D₃.
The data selector 22 selects the inputs D₀ - D₃ as an output Z_R in accordance with select signals S₀ and S₁ as indicated in the following table 1, and the output Z_R is fed into the signal selector 402 through the switch 14 as the interpolation data of the lacking part
.
The select signals S₀ and S₁ of the data selector 22 are formed by the level comparison portion 21.
More specifically, the comparator circuit 210 compares the output data R_{ℓ, n-1} of the latch circuit 10' and the read-out data R_{ℓ, n+1} of the line memory 902 and obtains the difference data ΔR_V:
which is compared with preset reference data D_S. The reference data D is set at about ½ of the maximum value of ΔR_V. Likewise, the comparator circuit 211 obtains the difference data ΔR_H between the output data items of the latch circuits 201 and 203, namely:
which is compared with the reference data D_S.
According to the results of the comparisons, the select signals S₀ and S₁ are determined as listed in the following table:
As understood from Table 2, when the difference data items ΔR_H and ΔR_V in the mode A are smaller than the reference data D_S, the level differences of the picture element data items around the lacking part X_{ℓ, n}are small, and accordingly, the red color hardly changes near this lacking part X_{ℓ, n}. In this case, the interpolation data D₀ similar to that of the embodiment in Fig. 11 is used. In the case of the mode B, it is expressed that the change between the picture element data items R_{ℓ+1, n} and R_{ℓ-1, n} of the two lines adjoining the lacking part X_{ℓ, n} is great and that the change between the picture element data items R_{ℓ, n-1} and R_{ℓ, n+1} holding the lacking part X_{ℓ, n} therebetween along the line L_ℓ is small. Accordingly, picture element data for this lacking part X_{ℓ, n} ought to be nearly equal to the two picture element data items R_{ℓ, n-1} and R_{ℓ, n+1} along the line L_ℓ. Therefore, the data selector 22 selects the output data A_V (= input D₁) of the mean value circuit 115, which is used as the interpolation data. This data item forms a vertical contour along the line L as to red. In the case of the mode C, the change between the picture element data items R_{ℓ, n-1} and R_{ℓ, n+1} of the same line L_ℓ holding the lacking part X_{ℓ, n} therebetween is great, and the change between the picture element data items R_{ℓ-1, n} and R_{ℓ+1, n} of the respective lines L_ℓ-1 and L_ℓ+1 is small. This signifies a horizontal contour concerning red, which is orthogonal to the line and passes through the picture element R_{ℓ, n}. In this case, the data selector 22 selects the mean value of the picture element data items R_{ℓ-1, n} and R_{ℓ+1, n} of small change, namely, the output data A_H (^{= D} ₂) of the mean value circuit 116, and this data item is used as the interpolation data.
In the case of the mode D, great changes are involved between the picture element data items R_{ℓ, n-1} and R _n+1 of the same line L_ℓ holding the lacking part X_{ℓ, n} therebetween and between the picture element data items R_{ℓ+1, n} and R_{ℓ-1, n} of the respective lines L_ℓ+1 and L_ℓ-1 holding the lacking part X_{ℓ, n} therebetween. This corresponds to a case where a contour concerning red exist in a direction oblique to the line. In this case, the data selector 22 selects the output data Z₂₃ (⁼ D₃) of the data selector 23 and supplies it to the signal selector 402 as the interpolation data. The output . data Z₂₃ of the data selector 23 will be explained below.
The reason why, in the preceding B and C modes, the mean value of the picture data items R_{ℓ+1, n} and R_{ℓ-1, n} and the mean value of the picture data items R_{ℓ, n-1} and R_{ℓ, n+1} of great change are not used as the interpolation data items, is that the digital R signal is of narrow band and has a small amount of information, so these mean values cannot always approximate the data items which ought to exist in the lacking parts X_{ℓ n}, namely, the picture element data items which have been missed by the synthesis switch 301 in Fig. la or 305 in Fig. lb. It is accordingly impossible also in the D mode that the mean value of the picture element data items R_{ℓ, n-1}, R_ℓ+1, n' R_ℓ, _n+1 and R_ℓ-1, n surrounding the lacking part X_{ℓ, n} be used as the interpolation data of this lacking part X_ℓ, n
On the other hand, the digital G signal is a wideband signal and does not lack any picture element data, so that it has a very large amount of information. Meanwhile, assuming that a contour exist obliquely on a frame thereby to give rise to the oblique contour concerning red as explained before, a similar oblique contour will exist concerning also the green color. Herein, when as to the digital R signal, the output data A, A or AS of the mean value circuit 115, 116 or 117 is used as the interpolation data of the lacking part X_ℓ, _n in this contour, the digital R signal changes gently in this contour part as indicated by a broken line in Fig.
14. This fact holds true for the digital B signal. In contrast, let it be assumed that, as illustrated in the figure, the digital G signal changes abruptly in this contour part, between the picture element data G_{ℓ, n-1} of the digital G signal corresponding to the picture element in which the digital R signal has the picture element data R_ℓ, _n-1 (hereinbelow, the same suffixes shall be assigned to the picture element data items of the respective digital color signals of an identical picture element) and the picture element data G_{ℓ, n}. Then, in the case where the aforementioned mean values are used as the interpolation data items for the lacking parts X_{ℓ, n} of the digital R and G signals as described above, the value of the picture data G_ℓ,n of the digital G signal is greater as compared with the values of the interpolation data items, and hence, the contour in the oblique direction becomes greenish on the printed frame.
As the contour, it is more preferable to be fringed in black rather than to be colored in green in this manner. To this end, in the case where the digital G signal changes as shown in Fig. 14, also the digital R signal should better be changed as indicated by a solid line, not by a dotted line. In this case, the picture element data R_{ℓ, n+1} is employed as the interpolation data of the lacking part Xe nof the digital R signal.
In this way, in the case of the mode D in Table 2 mentioned above, the interpolation data items of the lacking parts of the digital R and G signals are determined according to the change of the digital G signal. As this interpolation data item, the data selector 23 selects one of the picture element data R_{ℓ, n-1} from the latch circuit 10¹, the picture element data R_{ℓ, n+1} from the line memory 902 and the picture element data R_ℓ+1 n from the latch circuit 201 through select signals S₀' and S₁' in accordance with the change of the digital G signal.
There will now be explained the detection of the change of the digital G signal and the selecting operation of the data selector.23 responsive to this change.
The digital G signal read out from the frame memory 302 is supplied to the shift register portion 24, and one picture element data item is stored in each of the shift registers 241, 242 and 243. Here, the shift registers 241, 242 and 243 store the picture element data items of the digital G signal at the same picture element positions as those of data items which are being read out from the line memories 901, 902 and 903. Assuming that the read-out data from the line memory 902 be the picture element data R_{ℓ, n+1}, the picture element data G_{ℓ, +1} is being stored in the shift register 242. On this occasion, the latch circuit 252 latches the picture element data G_{ℓ, n} on the same line L_ℓ as that of the picture element data G_{ℓ, n+1} and preceding by one dot, and the latch circuit 26 latches the picture element data G_{ℓ, n-1} preceding by one more dot. In addition, the shift register 243 stores the picture element data G_{ℓ-1, n+1} of the line L_ℓ-1, and the latch circuit 253 latches the picture element data G_{ℓ-1, n} preceding by one dot. The positional relations of these picture element data items and the storing circuits are shown in Fig. 15 in correspondence with Fig. 13. The storing circuits are indicated by suffixes.
The level comparison portion 27 subjects the output data items of the latch circuits 252, 253 and - 26 to comparison processes similar to those of the level comparison portion 21.
More specifically, the comparator circuit 270 compares the output data items G_ℓ, n and G_{ℓ, n-1} of the respective latch circuits 252 and 26 and obtains difference data ΔG_V:
so as to decide the relation of magnitudes with the reference data D_S. The result of this decision is the S₀' input of the data selector 23. Besides, the comparator circuit 271 compares the output data items G_ℓ,n and G_{ℓ-1, n} of the respective latch circuits 252 and 253 and obtains difference data ΔG_H:
so as to decide the relation of magnitudes with the reference data D_S. The result of this decision is the S₁' input of the data selector 23. According to such S₀' and S₁' inputs, the data selector 23 selects any one of its inputs D₀' - D₃, as indicated in Table 3 below.
As seen from Fig. 15, ΔG_V denotes the change of picture element data items in the vertical direction (along the line), and ΔG_H the change of picture element data items in the horizontal direction.
In Table 3, in the case of the mode E, the horizontal and vertical changes of the picture element data items of the digital G signal are small in relation to the lacking part X_{ℓ, n} of the digital R signal. In this case, the digital G signal changes little and no contour exists as to this signal.
The data selector 23 selects the input D₀', namely, the picture element data R_ℓ, n-1 delivered by the latch circuit 10' and supplies it to the D₃ input end of the data selector 22. When both the inputs So and S₁of the data selector 22 are "1", that is, when the digital R signal has the oblique contour, the data selector 22 uses as the interpolation data of the lacking part X_ℓ, _n the picture element data R_{ℓ, n-1} on the same line L_ℓ and directly preceding the lacking part X_ℓ, _nand delivers it as an output Z_R. This corresponds to pre-holding in which the picture element data R_ℓ, _n-1 is held and is supplemented to the lacking part X_ℓ, n
The picture element data R_{ℓ, n-1}may well be replaced with picture element data R_{ℓ-1, n} as the interpolation data. This measure signifies pre-holding in the direction orthogonal to the line (namely, in the horizontal direction).
When the lacking part of the digital R signal is subjected to the data interpolation in this manner, the color change on the printed frame becomes abrupt to clarify the contour of the color.
In the mode F, the change of picture element data in the line direction (vertical direction) as to the digital G signal is great. In this case, the data selector 23 selects the picture element data R_{ℓ, n+1} (D₁' input) along the line L_ℓ of the lacking part X_{ℓ, n} of the digital R signal and directly succeeding the lacking part and delivers it to the data selector 22.. When both the inputs S₀ and S₁ are "1", the data selector 22 delivers the picture element data R_{ℓ, n+1} supplied from said data selector 23, as interpolation data for the lacking part X_n,ℓ. Thus, the contour concerning the digital R signal and the contour concerning the digital G signal are registered on the printed image, and they are made abrupt to equal extents so as to clarify the contours of the colors on the printed image. That is, the contours are not colored in green by way of example but become rather blackish.
In the mode G, picture element data items change greatly in the direction of adjacent lines (in the horizontal direction) as to the digital G signal. In this case, the data selector 23 selects the picture element data R_{ℓ+1, n} of the line L_ℓ+1 directly succeeding the line L_ℓ where the lacking part X_ℓ, _n exists, and it supplies the selected data to the data selector 22: When both the inputs S₀ and S₁ are "1", the data selector 22 delivers the picture element data R_{ℓ+1, n} supplied from the data selector 23, as interpolation data for the lacking part x _n. Thus, the same effects as in the mode F are attained.
The mode H is a case where the oblique contour exists concerning the digital G signal. In this case, as in the case of the mode G, the picture element data R_ℓ+1, n (the D₃ ¹ input of the data selector 23) as the interpolation data of the lacking part X_{ℓ, n} of the digital R signal. This is intended to attain the same effects as in the mode F, and the picture element data R_ℓ+1, _n may well be replaced with the picture element data R_{ℓ, n+1} directly succeeding the lacking part X_{ℓ, n} of the line L_ℓ.
As described above, in the case where the lacking part X_{ℓ n} of the digital R signal registers with the oblique contour, the interpolation data of this lacking part X_ℓ, is set according to the change of the digital G signal which has the large amount of information with no data omitted, whereby the contour in the printed image becomes clear without being colored. Besides, contours in the horizontal and vertical directions concerning the digital R signal can be clearly reproduced. Needless to say, the data interpolation processing stated above is similarly executed for the digital B signal.
Although, in this practicable example, the processes of holding and operating various data items and the function of selecting data have been realized with hardware, they may well be realized with software by the use of a microcomputer or the like.
In the level comparison portions 21 and 27, the items of reference data D_S have been equalized and have been set at ½ of the dynamic range. Needless to say, however, the reference data D_S can be set at will, in accordance with a printing function or for each of the level comparison portions 21 and 27 or the comparator circuits 210, 211, 270 and 271,
Further, it is a matter of course that, as in the preceding practicable examples, the first and last lacking parts of each line may be supplemented or that the first and last dots of each line and the first and last lines may be similarly made white on the printing paper.
Although, in the embodiments, the R and B signals are multiplexed and stored in the identical frame memory, they may well be stored in individual memories, and if necessary, the sampling rate may well be lowered for only the B signal by way of example so as to reduce the capacity of the memory.

Claims

1. A signal processing circuit for a color video printer wherein images of first, second and third colors are successively printed to reproduce a color print on a surface for printing, comprising:

a signal input portion which includes first, second and third color signal input ends that are respectively supplied with first, second and third color signals corresponding to the images of the first, second and third colors,

memory means which includes a first memory for storing the first color signal, and a second memory for storing the second and third color signals,

switch means which is connected to said second and third color signal input ends and said second memory, and which supplies said second memory with the second and third color signals alternately,

data interpolation means which has its input end connected to said second memory, and which generates an interpolation data signal of the second color signal or the third color signal stored in said second memory, to deliver a fourth color signal with the second or third color signal interpolated by the interpolation data signal,

selector means which has its input ends connected to said first memory and said data interpolation means, and which delivers the first color signal read out from said first memory and the fourth color signal delivered from said data interpolation means, as print signals in a printing order, and

print means which has its input end connected to said selector means, and which prints the images on the surface-for-printing successively in accordance with the print signals supplied from said selector means.

2. A signal processing circuit for a color video printer as defined in Claim 1, wherein said data interpolation means comprises detection means to detect a signal level of the second color signal, and contour data generation means to deliver a color signal preceding by one picture element, as interpolation data in accordance with the output of said detection means.

3. A signal processing circuit for a color video printer as defined in Claim 1, wherein said data interpolation means delivers data which is previously determined as data of a first line or a last line.

4. A signal processing circuit for a color video printer as defined in Claim 3, wherein the previously determined data is data which reproduces a white image.

5. A signal processing circuit for a color video printer as defined in Claim 1, wherein said data interpolation means comprises a first line memory which holds data D_2L-1 of a (2L - l)-th line, a second line memory which holds data D_2L of a 2L-th line, a third line memory which holds data D_2L+l of a (2L + l)-th line, and interpolation data generation means to create interpolation data D_{2L, 2n-1} of an n-th address for the data D_2L of the 2L-th line, on the basis of the data items D_2L-1, ^D _2L and ^D _2L+1 held in said first, second and third line memories respectively, as follows:

6. A signal processing circuit for a color video printer wherein images of a plurality of colors are successively printed on a surface for printing so as to reproduce a color print, comprising:

a color signal input protion which includes a plurality of color signal input ends that are respectively supplied with a plurality of analog color signals corresponding to the plurality of colors,

first analog-to-digital conversion means which converts a first one of the plurality of analog color signals into a first digital color signal at a first sampling rate,

second analog-to-digital conversion means which converts a second one of the plurality of analog color signals into a second digital color signal at a second sampling rate,

first memory means which is connected to said first analog-to-digital conversion means and which stores the first digital color signal,

second memory means which is connected to said second analog-to-digital conversion means and which stores the second digital color signal,

data interpolation-means which has its input end connected to said second memory, and which generates an interpolation data signal of the second digital color signal stored in said second memory, to deliver a third digital color signal with the second digital color signal interpolated by the interpolation data signal,

selector means which has its input ends connected to said first memory and said data interpolation means, and which delivers the first digital color signal read out from said first memory and the third digital color signal delivered from said data interpolation means, as print signals in a printing order, and

print means which has its input end connected to said selector means, and which prints the images of the plurality of colors on the surface-for-printing in accordance with the print signals supplied from said selector means.

7. A signal processing circuit for a color video printer as defined in Claim 6, wherein the second sampling rate is not higher than at most 2 of the first sampling rate.

8. A signal processing circuit for a color video printer as defined in Claim 7, wherein said data interpolation means comprises detection means to detect a signal level of the first digital color signal, and contour data generation means to deliver a color signal preceding by one picture element, as interpolation data in accordance with the output of said detection means.

9. A signal processing circuit for a color video signal as defined in Claim 7, wherein said data interpolation means delivers data which is previously determined as data of a first line or a last line.

10. A signal processing circuit for a color video signal as defined in Claim 9, wherein the previously determined data is data which reproduces a white image.

11. A signal processing circuit for a color video signal as defined in Claim 6, wherein said data interpolation means comprises a first line memory which holds data D_2L-1 of a (2L - l)-th line, a second line memory which holds data D_2L of a 2L-th line, a third line memory which holds data D_2L+l of a (2L + l)-th line, and interpolation data generation means to create interpolation data D_2L, _2n-1 of an n-th address for the data D_2L of the 2L-th line, on the bais of the data items D2L-1, D_2L and ^D _2L+1 held in said first, second and third line memories respectively, as follows:

12. A signal processing circuit for a color video printer having memories which store a plurality of color signals respectively, and print means which reads out the color signals from the memories in succession so as to print images on a printing paper in accordance with the color signals read out, comprising:

memory control means which samples at least one of the plurality of color signals so as to store it in said memory, and

data interpolation means which has its input end connected to said memory and its output end connected to said print means, and which generates data of unsampled points of the color signal sampled and stored in said memory.

13. A signal processing circuit for a color video printer as defined in Claim 12, wherein said memory control means comprises an analog color signal input end, and analog-to-digital conversion means having its input end connected to said analog color signal input end and its output end connected to said memory and to convert an analog color signal applied to said analog signal input end, into a digital color signal and to deliver the latter to said memory.

14. A signal processing circuit for a color video printer as defined in Claim 13, wherein said analog-to-digital conversion means comprises a plurality of analog-to-digital converters which convert a plurality of analog color signals into digital color signals respectively.

15. A signal processing circuit for a color video printer as defiend in Claim 14, wherein said analog-to-digital conversion means comprises switch . means connected to output ends of at least two of said plurality of analog-to-digital converters, and to select and deliver one of the plurality of digital color signals applied thereto.

16. A signal processing circuit for a color video printer as defined in Claim 14, wherein said analog-to-digital conversion means comprises switch means having its input ends connected to at least two analog signal input ends and its output end connected to one of said plurality of analog-to-digital converters, and to select and deliver one of a plurality of analog colors signals applied thereto.

17. A signal processing circuit for a color video printer as defined in Claim 12, wherein said data interpolation means comprises detection means to detect signal levels of unsampled color signals of said plurality of color signals, and contour data generation means to deliver color signals preceding by one picture element, as interpolation data in accordance with the outputs of said detection means.

18. A signal processing circuit for a color video signal as defined in Claim 12, wherein said data interpolation means comprises a first line memory which holds data D_2L-1 of a (2L - l)-th line, a second line memory which holds data D_2L of a 2L-th line, a third line memory which holds data D_2L+l of a (2L + l)-th line, and interpolation data generation means to create interpolation data D_{2L, 2n-1} of an n-th address for the data D_2L of the 2L-th line, on the bais of the data items D_2L-1, ^D _2L and ^D _2L+1 held in said first, second and third line memories respectively, as follows:

19. A signal processing circuit for a color video signal ad defined in Claim 12, wherein said data interpolation means delivers data which is previously determined as data of a first line or a last line.

20. A signal processing circuit for a color video signal as defined in Claim 19, wherein the previously determined data is data which reproduces a white image.

21. A signal processing circuit for a color video printer as defined in Claim 12, wherein said data interpolation means delivers data for reproducing a white image, as data of a first line or a last line.